In the field of soft robotics, physical agents are built to conform to their surroundings and mimic biological actuation. One of the key focuses of soft robotics is on controlling kinematics and dynamics through material properties to enable interactions with environments. It is a design problem to find the form which can achieve the desired dynamic behavior.

Evolving Soft Bodies explores the use of genetic algorithms to arrive at the configurational geometry for a desired behavior. The soft body system consisting of nodes such as air pockets is envisioned to be controlled with a single pneumatic actuator. This explores a top-down approach to find the desired geometry. The soft body is simulated as a spring system and the behavior of traversing a plane ground has been adapted by using genetic algorithms.